Face Image Reflection Removal
- PDF / 3,228,144 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 22 Downloads / 225 Views
Face Image Reflection Removal Renjie Wan1
· Boxin Shi2,3
· Haoliang Li1 · Ling-Yu Duan2,3 · Alex C. Kot1
Received: 13 December 2019 / Accepted: 14 August 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Face images captured through glass are usually contaminated by reflections. The low-transmitted reflections make the reflection removal more challenging than for general scenes because important facial features would be completely occluded. In this paper, we propose and solve the face image reflection removal problem. We recover the important facial structures by incorporating inpainting ideas into a guided reflection removal framework, which takes two images as the input and considers various face-specific priors. We use a newly collected face reflection image dataset to train our model and compare with state-of-the-art methods. The proposed method shows advantages in estimating reflection-free face images for improving face recognition. Keywords Reflection removal · Deep learning · Face images · Optical flow
1 Introduction As one of the commonly observed subjects in computer vision, face images are often captured by various types of Communicated by Yoichi Sato. The research work was done at the Rapid-Rich Object Search (ROSE) Lab, Nanyang Technological University. The work is supported in part by the Wallenberg-NTU Presidential Postdoctoral Fellowship, the NTU-PKU Joint Research Institute, a collaboration between the Nanyang Technological University and Peking University that is sponsored by a donation from the Ng Teng Fong Charitable Foundation, and the Science and Technology Foundation of Guangzhou Huangpu Development District under Grant 201902010028. This research is in part supported by the National Natural Science Foundation of China under Grants 61872012 and U1611461, and Beijing Academy of Artificial Intelligence (BAAI). Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11263-020-01372-5) contains supplementary material, which is available to authorized users.
B B
Renjie Wan [email protected] Boxin Shi [email protected]
1
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, Singapore
2
National Engineering Laboratory for Video Technology, Department of CS, Peking University, Beijing, China
3
Peng Cheng Laboratory, Shenzhen, China
imaging sensors under unconstrained wild scenarios, which bring different types of distortions to the clear face images. When face images are captured behind a piece of glass, the reflection-contaminated face images not only unpleasantly affect the human perception but also degrade the performance of vision algorithms applied on the face. Therefore, it is of great interest to remove the reflections and enhance the visibility of the human faces behind glass. Different from general objects or scenes, faces have their own specific priors awarded by humans, and a slight reflection (transmitted) distortion may significantly affect human perception (Liu
Data Loading...
